[SOLVED] Two BSoDs in two weeks - hal.dll, ntoskrnl.exe

[jelcher]

Hello, I built my new PC almost 2 months ago. Since then no problems, no BSODs and generally everything was fine. First BSoD happened suddenly when I was playing Resident Evil 6 two weeks ago. After that I did sfc /scannow, chkdsk and Windows Memory Diagnostic, and they didn't give me any errors. I also increased my pagefile.sys, because I thought I was running out of memory. Sadly I don't have .dmp file, but I remember it was Page fault in nonpaged area caused by ntoskrnl.exe.

Second BSoD happened yesterday. I was changing some settings in Realtek Control Panel (because my speakers weren't playing good, probably irrelevant), also turned off Telemetry service. After that I trying to reboot my PC but after choosing reboot optiong - bang, BSoD. System Service Exception caused by hal.dll this time. Link to .dmp: https://www29.zippyshare.com/v/z2RuSSxI/file.html

My question is - where is a cause? After second BSoD I did scans mentioned before again and again no errors. I also updated my drivers for sound card, network card, chipset and SATA controller as well as turned off XMP profile (from 3000 MHz to 2133 MHz). I ran a memtest and there is a report: https://www29.zippyshare.com/v/SnJ4XoAE/file.html Only errors was during pass #4 when doing hammer test, another tests are passed. Sadly testing one RAM stick at time may be difficult for me (desk is very close to the wall). Temps are normal, rarely exceeding 60C. When writing that post, it was 30C for CPU and 34C for GPU. Nothing is overclocked.

My specs:
CPU: Intel Core i5-8600K 3.60GHz
GPU: Gigabyte GeForce GTX 1060 Windforce OC 6GB GDDR5
RAM: Corsair Vengeance LPX, DDR4, 16 GB, 3000MHz, CL15
MoBo: Z370 AORUS Gaming K3, BIOS: F11 10/30/2018
OS: Windows 10 Education
PSU: SeaSonic Focus Plus Gold 550W

 

[Colif]

Download and/or run Gigabyte App Center from utilities here - Link

see if it suggests any new drivers

ntoskrnl = windows kernel. it is the brains of windows, it It handles all driver requests, power management, and memory management. It sits between Hardware and Applications. It got blamed but its not the cause

hal.dll = hardware abstraction layer. It sits between windows and hardware (actually between ntoskrnl and hardware)

you may want to do the hassle of testing ram 1 stick at a time as doing both, you don't know which stick has the hammer test errors. if you keep getting random new errors, it could well be a bad stick

 

[jelcher]

There are driver updates, but mostly it suggests me drivers older than these installed yesterday (which were downloaded from mobo website).

So, you think that RAM is a culprit, not changing settings before reboot? I read that hammer test fails are rarely causing BSoDs...

 

[Colif]

i will ask a user called @gardenman to convert dump files, we see what they have to say.

I don't blame ram, but I have answered enough BSOD posts to know better than to not suggest it. It always seems to be the cause when I forget to suggest it.

 

[jelcher]

Not sure if it's important but today Chrome crashed two tabs while regular browsing (first 3 hours ago, second now) with "Aw, Snap!" error. I hadn't this error before.

Additionally I scanned my PC with Malwarebytes earlier today. No threats.

And another BSoD - Kernel Security Check Failure: https://www113.zippyshare.com/v/Y5GNenil/file.html

 

[gamerk316]

If ANY test in Memtest86+ failed, then your entire system is suspect. Bad RAM will cause the exact issues you are seeing. This also explains Chrome crashing (Chrome using a bad memory address, but not crashing the system).

You are going to have to replace at least one stick of RAM; you can either test each stick individually to isolate the bad one, or just buy a new set (I always use matched pairs personally; less headaches).

 

[jelcher]

Even the latest BSoD suggests RAM fault?
Also, I used memtest86, not memtest86+. Memtest86+ doesn't recognize my RAM, because it's DDR4. However I think my RAM should pass memtest86+, because it passed all tests except hammer test while testing with memtest86, which isn't in memtest86+

 

[gamerk316]

The last BSOD basically proves it as far as I'm concerned; that's not a BSOD that should ever come up. The fact it did basically proves serious memory corruption is occurring.

Either your RAM settings/voltages are incorrect, or the RAM itself is faulty. Either case could likely explain the errors and BSODs you are describing.

 

[jelcher]

So, what I gonna do? Test RAM again, this time only one stick at time?

 

[gardenman]

Hi, I ran the dump files through the debugger and got the following information: https://pste.eu/p/NwjT.html

Spoiler

File information:	021919-7734-01.dmp (Feb 19 2019 - 07:04:26)
Bugcheck:	KERNEL_SECURITY_CHECK_FAILURE (139)
Probably caused by:	memory_corruption (Process: RuntimeBroker.exe)
Uptime:	0 Day(s), 1 Hour(s), 21 Min(s), and 15 Sec(s)

File information:	021819-8031-01.dmp (Feb 18 2019 - 10:59:29)
Bugcheck:	SYSTEM_SERVICE_EXCEPTION (3B)
Probably caused by:	memory_corruption (Process: WindowsInternal.ComposableShell.Experiences.TextInput.InputApp)
Uptime:	5 Day(s), 3 Hour(s), 26 Min(s), and 27 Sec(s)

Possible Motherboard page: https://www.gigabyte.com/Motherboard/Z370-AORUS-Gaming-K3-rev-10#kf
If I got the motherboard page right, there is a BIOS update available for your system. Wait for additional information before deciding to update or not. Important: Verify that I have linked to the correct motherboard. Updating your BIOS can be risky. Never try it when you might lose power (lightning storms, recent power outages, etc).

I can't help you with this. Someone else will post with more information soon. Please wait for more answers. Good luck.

 

[jelcher]

Just updated BIOS and loaded its default settings. Will see if it helps.

 

[jelcher]

I just looked into Reliability Monitor and saw that:

1. just before first "Aw, snap!" error "IP Helper" service crashed
2. "WindowsInternal.ComposableShell.Experiences.TextInput.InputApp" (process which caused BSoD from yesterday when rebooting) crashed today when I tried type an emoji (shortcut: Windows+Period)
3. just before BSoD from today when I tried type into finder from Start Menu "Microsoft Network Realtime Inspection Service" crashed twice

 

[Colif]

these aren't normal processes to crash... never seen that 2nd one crash before.
see if this helps

right click start button
choose powershell (admin)
type SFC /scannow and press enter
once its completed, copy/paste this command into same window:
Repair-WindowsImage -Online -RestoreHealth and press enter
SFC fixes system files, second command cleans image files, re run SFC if it failed to fix all files and restart PC

Check ram as per above, the fact it isn't a normal process to crash could mean it is ram

 

[jelcher]

SFC and DISM done, no errors.
RAM will be checked again this weekend probably.
I'll signal if new BSoD happens.

 

[jelcher]

So, I ran driver verifier (suggested on another forum), but after every logging in PC just rebooted itself (without BSoD). So I restored from restore point and after successfully finish there was another BSOD - System Thread Exception Not Handled. Sadly there is no minidump again, despite proper settings. After reboot it works normally - except Chrome, which crashing all the time. MS Edge works.
After reinstall Chrome works again.

 

[gamerk316]

Yeah, this certainly sounds like RAM to me. I'd run a memory checker ASAP and see if anything crops up.

 

[jelcher]

So, @gamerk316 was right. RAM is faulty. One stick is fine, no errors, but other has hundreds of errors in first test. Sticks were tested in same DIMM slot, so mobo is fine too. I'm gonna RMA my RAM. Shame that I must return both sticks, because they're from same set.